{"source":"AIMap","schema":"benchmark/v1","updated":"2026-10-22","benchmark":{"id":"gpqa-diamond","name":"GPQA Diamond","category":"science","unit":"%","higherIsBetter":true,"status":"active","sources":["epoch"],"primaryVariant":"epoch","url":"https://github.com/idavidrein/gpqa","year":2023,"desc":{"en":"198 graduate-level multiple-choice questions in biology, physics and chemistry written to be 'Google-proof'.","zh-hans":"198 道研究生水平的生物、物理、化学选择题，设计为“搜不到答案”。"},"measures":{"en":"Expert-level scientific reasoning. Scores here are Epoch AI's own runs (accuracy, best across scorers).","zh-hans":"专家级科学推理。本站分数来自 Epoch AI 自己的统一评测（准确率）。"},"limits":{"en":"Only 198 questions, so ±3 points is noise; random guessing gives 25%; top models are close to saturation.","zh-hans":"只有 198 题，3 个百分点以内的差距基本是噪声；随机猜答案也有 25%；头部模型接近饱和。"},"contamination":"medium","contaminationNote":{"en":"Questions are public since 2023.","zh-hans":"题目自 2023 年起公开。"},"stats":{"results":282,"models":217,"latest":"2026-10-19","bySourceType":{"independent":282}}},"sources":[{"id":"epoch","name":"Epoch AI Benchmarking Hub","license":"CC BY 4.0","policy":"open","attribution":"Epoch AI, 'Capabilities & benchmarking', https://epoch.ai/benchmarks (CC BY 4.0)","site":"https://epoch.ai/benchmarks"}],"results":[{"model":"gemini-4-argon","modelName":"Gemini 4 Argon","provider":"Google","rawName":"Gemini 4 Argon","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":96.4,"unit":"%","rank":1,"rankComputedBySite":true,"date":"2026-10-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-22"},{"model":"gpt-6-astra","modelName":"GPT-6 Astra","provider":"OpenAI","rawName":"GPT-6 Astra","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":95.77,"unit":"%","rank":2,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-5-5","modelName":"Claude Sonnet 5.5","provider":"Anthropic","rawName":"Claude Sonnet 5.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":95.58,"unit":"%","rank":3,"rankComputedBySite":true,"date":"2026-09-29","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-6-1-sol","modelName":"GPT-6.1 Sol","provider":"OpenAI","rawName":"GPT-6.1 Sol","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":95.39,"unit":"%","rank":4,"rankComputedBySite":true,"date":"2026-09-29","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-8-flash","modelName":"Gemini 3.8 Flash","provider":"Google","rawName":"Gemini 3.8 Flash","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":95.39,"unit":"%","rank":4,"rankComputedBySite":true,"date":"2026-09-02","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-7-flash","modelName":"Gemini 3.7 Flash","provider":"Google","rawName":"Gemini 3.7 Flash","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":94.82,"unit":"%","rank":6,"rankComputedBySite":true,"date":"2026-08-14","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4-pro","modelName":"GPT-5.4 Pro","provider":"OpenAI","rawName":"GPT-5.4 Pro","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":94.6,"unit":"%","rank":7,"rankComputedBySite":true,"date":"2026-03-20","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-1-pro","modelName":"Gemini 3.1 Pro","provider":"Google","rawName":"Gemini 3.1 Pro","variant":"epoch","setting":"Epoch AI run, accuracy","score":94.44,"unit":"%","rank":8,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-6-sol","modelName":"GPT-6 Sol","provider":"OpenAI","rawName":"GPT-6 Sol","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":94.32,"unit":"%","rank":9,"rankComputedBySite":true,"date":"2026-09-22","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"grok-4-6","modelName":"Grok 4.6","provider":"xAI","rawName":"Grok 4.6","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":94,"unit":"%","rank":10,"rankComputedBySite":true,"date":"2026-08-12","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-5","modelName":"GPT-5.5","provider":"OpenAI","rawName":"GPT-5.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":94,"unit":"%","rank":10,"rankComputedBySite":true,"date":"2026-04-24","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-5-pro","modelName":"GPT-5.5 Pro","provider":"OpenAI","rawName":"GPT-5.5 Pro","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":93.92,"unit":"%","rank":12,"rankComputedBySite":true,"date":"2026-04-24","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-5","modelName":"Claude Opus 5","provider":"Anthropic","rawName":"Claude Opus 5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":93.88,"unit":"%","rank":13,"rankComputedBySite":true,"date":"2026-07-24","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-6-sol","modelName":"GPT-5.6 Sol","provider":"OpenAI","rawName":"GPT-5.6 Sol","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":93.5,"unit":"%","rank":14,"rankComputedBySite":true,"date":"2026-07-09","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"grok-4-5","modelName":"Grok 4.5","provider":"xAI","rawName":"Grok 4.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":93.43,"unit":"%","rank":15,"rankComputedBySite":true,"date":"2026-07-08","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-6-terra","modelName":"GPT-5.6 Terra","provider":"OpenAI","rawName":"GPT-5.6 Terra","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":93.31,"unit":"%","rank":16,"rankComputedBySite":true,"date":"2026-07-09","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4","modelName":"GPT-5.4","provider":"OpenAI","rawName":"GPT-5.4","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":93.3,"unit":"%","rank":17,"rankComputedBySite":true,"date":"2026-03-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v4-1-flash-1017","modelName":"DeepSeek V4.1 Flash 1017","provider":"DeepSeek","rawName":"DeepSeek V4.1 Flash 1017","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":93.18,"unit":"%","rank":18,"rankComputedBySite":true,"date":"2026-10-19","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-22"},{"model":"grok-4-6","modelName":"Grok 4.6","provider":"xAI","rawName":"Grok 4.6","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":93.18,"unit":"%","rank":18,"rankComputedBySite":true,"date":"2026-08-14","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"kimi-k3","modelName":"Kimi K3","provider":"Moonshot AI","rawName":"Kimi K3","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":93.12,"unit":"%","rank":20,"rankComputedBySite":true,"date":"2026-07-16","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-5","modelName":"Claude Opus 5","provider":"Anthropic","rawName":"Claude Opus 5","variant":"epoch","setting":"Epoch AI run, accuracy","score":92.93,"unit":"%","rank":21,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"grok-4-7","modelName":"Grok 4.7","provider":"xAI","rawName":"Grok 4.7","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":92.74,"unit":"%","rank":22,"rankComputedBySite":true,"date":"2026-09-22","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen-3-8-max","modelName":"Qwen 3.8 Max","provider":"Alibaba (Qwen)","rawName":"Qwen 3.8 Max","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":92.68,"unit":"%","rank":23,"rankComputedBySite":true,"date":"2026-08-04","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-pro","modelName":"Gemini 3 Pro","provider":"Google","rawName":"Gemini 3 Pro","variant":"epoch","setting":"Epoch AI run, accuracy","score":92.61,"unit":"%","rank":24,"rankComputedBySite":true,"date":"2025-11-19","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-8-max-0902","modelName":"Qwen3.8 Max (0902)","provider":"Alibaba (Qwen)","rawName":"Qwen3.8 Max (0902)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":92.3,"unit":"%","rank":25,"rankComputedBySite":true,"date":"2026-09-02","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"kimi-k3","modelName":"Kimi K3","provider":"Moonshot AI","rawName":"Kimi K3","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":91.92,"unit":"%","rank":26,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v4-pro-0813","modelName":"DeepSeek V4 Pro 0813","provider":"DeepSeek","rawName":"DeepSeek V4 Pro 0813","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":91.67,"unit":"%","rank":27,"rankComputedBySite":true,"date":"2026-08-18","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-6-luna","modelName":"GPT-5.6 Luna","provider":"OpenAI","rawName":"GPT-5.6 Luna","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":91.6,"unit":"%","rank":28,"rankComputedBySite":true,"date":"2026-07-09","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":91.4,"unit":"%","rank":29,"rankComputedBySite":true,"date":"2025-12-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v4-flash-0731","modelName":"DeepSeek V4 Flash 0731","provider":"DeepSeek","rawName":"DeepSeek V4 Flash 0731","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":91.04,"unit":"%","rank":30,"rankComputedBySite":true,"date":"2026-08-02","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"glm-5-3","modelName":"GLM-5.3","provider":"Zhipu AI (Z.ai)","rawName":"GLM-5.3","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":90.91,"unit":"%","rank":31,"rankComputedBySite":true,"date":"2026-08-24","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-7-max","modelName":"Qwen3.7-Max","provider":"Alibaba (Qwen)","rawName":"Qwen3.7-Max","variant":"epoch","setting":"Epoch AI run, accuracy","score":90.91,"unit":"%","rank":31,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v4-pro","modelName":"DeepSeek-V4-Pro","provider":"DeepSeek","rawName":"DeepSeek-V4-Pro","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":90.91,"unit":"%","rank":31,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"kimi-k2-6","modelName":"Kimi K2.6","provider":"Moonshot AI","rawName":"Kimi K2.6","variant":"epoch","setting":"Epoch AI run, accuracy","score":90.78,"unit":"%","rank":34,"rankComputedBySite":true,"date":"2026-05-01","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-5","modelName":"GPT-5.5","provider":"OpenAI","rawName":"GPT-5.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":90.66,"unit":"%","rank":35,"rankComputedBySite":true,"date":"2026-05-05","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-5-5","modelName":"Claude Opus 5.5","provider":"Anthropic","rawName":"Claude Opus 5.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":90.59,"unit":"%","rank":36,"rankComputedBySite":true,"date":"2026-09-22","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-5","modelName":"Claude Sonnet 5","provider":"Anthropic","rawName":"Claude Sonnet 5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":90.53,"unit":"%","rank":37,"rankComputedBySite":true,"date":"2026-07-01","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 32k thinking","score":90.53,"unit":"%","rank":37,"rankComputedBySite":true,"date":"2026-02-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-6-luna","modelName":"GPT-6 Luna","provider":"OpenAI","rawName":"GPT-6 Luna","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":90.47,"unit":"%","rank":39,"rankComputedBySite":true,"date":"2026-09-22","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"glm-5-3-flash","modelName":"GLM-5.3-Flash","provider":"Zhipu AI (Z.ai)","rawName":"GLM-5.3-Flash","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":90.15,"unit":"%","rank":40,"rankComputedBySite":true,"date":"2026-08-26","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-7","modelName":"Claude Opus 4.7","provider":"Anthropic","rawName":"Claude Opus 4.7","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":90.15,"unit":"%","rank":40,"rankComputedBySite":true,"date":"2026-04-17","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"glm-5-1","modelName":"GLM-5.1","provider":"Zhipu AI (Z.ai)","rawName":"GLM-5.1","variant":"epoch","setting":"Epoch AI run, accuracy","score":89.9,"unit":"%","rank":42,"rankComputedBySite":true,"date":"2026-08-10","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-6-sol","modelName":"GPT-5.6 Sol","provider":"OpenAI","rawName":"GPT-5.6 Sol","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":89.9,"unit":"%","rank":42,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4","modelName":"GPT-5.4","provider":"OpenAI","rawName":"GPT-5.4","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":89.9,"unit":"%","rank":42,"rankComputedBySite":true,"date":"2026-07-15","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v4-1-flash","modelName":"DeepSeek V4.1 Flash","provider":"DeepSeek","rawName":"DeepSeek V4.1 Flash","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":89.84,"unit":"%","rank":45,"rankComputedBySite":true,"date":"2026-10-09","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"muse-spark","modelName":"Muse Spark","provider":"Meta","rawName":"Muse Spark","variant":"epoch","setting":"Epoch AI run, accuracy","score":89.8,"unit":"%","rank":46,"rankComputedBySite":true,"date":"2026-04-08","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-haiku-5-5","modelName":"Claude Haiku 5.5","provider":"Anthropic","rawName":"Claude Haiku 5.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":89.65,"unit":"%","rank":47,"rankComputedBySite":true,"date":"2026-10-09","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v4-pro","modelName":"DeepSeek-V4-Pro","provider":"DeepSeek","rawName":"DeepSeek-V4-Pro","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":89.65,"unit":"%","rank":47,"rankComputedBySite":true,"date":"2026-06-16","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-flash","modelName":"Gemini 3 Flash","provider":"Google","rawName":"Gemini 3 Flash","variant":"epoch","setting":"Epoch AI run, accuracy","score":89.39,"unit":"%","rank":49,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"grok-4-20","modelName":"Grok 4.20","provider":"xAI","rawName":"Grok 4.20","variant":"epoch","setting":"Epoch AI run, accuracy","score":89.33,"unit":"%","rank":50,"rankComputedBySite":true,"date":"2026-07-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-5-flash","modelName":"Gemini 3.5 Flash","provider":"Google","rawName":"Gemini 3.5 Flash","variant":"epoch","setting":"Epoch AI run, accuracy","score":88.89,"unit":"%","rank":51,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4","modelName":"GPT-5.4","provider":"OpenAI","rawName":"GPT-5.4","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":88.89,"unit":"%","rank":51,"rankComputedBySite":true,"date":"2026-07-15","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"grok-4-3-beta","modelName":"Grok 4.3 Beta","provider":"xAI","rawName":"Grok 4.3 Beta","variant":"epoch","setting":"Epoch AI run, accuracy","score":88.83,"unit":"%","rank":53,"rankComputedBySite":true,"date":"2026-06-17","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 64k thinking","score":88.76,"unit":"%","rank":54,"rankComputedBySite":true,"date":"2026-02-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"muse-spark-open","modelName":"Muse Spark Open","provider":"Meta","rawName":"Muse Spark Open","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":88.57,"unit":"%","rank":55,"rankComputedBySite":true,"date":"2026-10-15","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-22"},{"model":"inkling-small","modelName":"Inkling-Small","provider":"Thinking Machines","rawName":"Inkling-Small","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":88.51,"unit":"%","rank":56,"rankComputedBySite":true,"date":"2026-08-14","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen-3-6-plus","modelName":"Qwen 3.6 Plus","provider":"Alibaba (Qwen)","rawName":"Qwen 3.6 Plus","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 2026-04-02","score":88.38,"unit":"%","rank":57,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-6","modelName":"Claude Opus 4.6","provider":"Anthropic","rawName":"Claude Opus 4.6","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":88.38,"unit":"%","rank":57,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"inkling","modelName":"Inkling","provider":"Thinking Machines","rawName":"Inkling","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":88.26,"unit":"%","rank":59,"rankComputedBySite":true,"date":"2026-08-05","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":88.19,"unit":"%","rank":60,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-large-4","modelName":"Mistral Large 4","provider":"Mistral AI","rawName":"Mistral Large 4","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":88.07,"unit":"%","rank":61,"rankComputedBySite":true,"date":"2026-10-12","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-22"},{"model":"kimi-k2-7-code","modelName":"Kimi K2.7 Code","provider":"Moonshot AI","rawName":"Kimi K2.7 Code","variant":"epoch","setting":"Epoch AI run, accuracy","score":87.88,"unit":"%","rank":62,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-7-plus","modelName":"Qwen3.7-Plus","provider":"Alibaba (Qwen)","rawName":"Qwen3.7-Plus","variant":"epoch","setting":"Epoch AI run, accuracy","score":87.88,"unit":"%","rank":62,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-5","modelName":"Claude Opus 5","provider":"Anthropic","rawName":"Claude Opus 5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":87.88,"unit":"%","rank":62,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":87.88,"unit":"%","rank":62,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"glm-5","modelName":"GLM-5","provider":"Zhipu AI (Z.ai)","rawName":"GLM-5","variant":"epoch","setting":"Epoch AI run, accuracy","score":87.82,"unit":"%","rank":66,"rankComputedBySite":true,"date":"2026-02-12","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-1","modelName":"GPT-5.1","provider":"OpenAI","rawName":"GPT-5.1","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":87.63,"unit":"%","rank":67,"rankComputedBySite":true,"date":"2025-11-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"kimi-k2-5","modelName":"Kimi K2.5","provider":"Moonshot AI","rawName":"Kimi K2.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Fireworks","score":87.6,"unit":"%","rank":68,"rankComputedBySite":true,"date":"2026-02-02","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen-3-6-max-preview","modelName":"Qwen 3.6 Max (Preview)","provider":"Alibaba (Qwen)","rawName":"Qwen 3.6 Max (Preview)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Preview","score":87.37,"unit":"%","rank":69,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-6-terra","modelName":"GPT-5.6 Terra","provider":"OpenAI","rawName":"GPT-5.6 Terra","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":87.37,"unit":"%","rank":69,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-6","modelName":"Claude Sonnet 4.6","provider":"Anthropic","rawName":"Claude Sonnet 4.6","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 32k thinking","score":87.37,"unit":"%","rank":69,"rankComputedBySite":true,"date":"2026-02-20","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"grok-4","modelName":"Grok 4","provider":"xAI","rawName":"Grok 4","variant":"epoch","setting":"Epoch AI run, accuracy","score":87,"unit":"%","rank":72,"rankComputedBySite":true,"date":"2025-07-09","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4-mini","modelName":"GPT-5.4 Mini","provider":"OpenAI","rawName":"GPT-5.4 Mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: xhigh","score":86.87,"unit":"%","rank":73,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-7","modelName":"Claude Opus 4.7","provider":"Anthropic","rawName":"Claude Opus 4.7","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":86.36,"unit":"%","rank":74,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":86.17,"unit":"%","rank":75,"rankComputedBySite":true,"date":"2025-10-29","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 32k thinking","score":86.05,"unit":"%","rank":76,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-6-27b","modelName":"Qwen3.6 27B","provider":"Alibaba (Qwen)","rawName":"Qwen3.6 27B","variant":"epoch","setting":"Epoch AI run, accuracy","score":85.86,"unit":"%","rank":77,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-5-397b-a17b","modelName":"Qwen3.5 397B-A17B","provider":"Alibaba (Qwen)","rawName":"Qwen3.5 397B-A17B","variant":"epoch","setting":"Epoch AI run, accuracy","score":85.86,"unit":"%","rank":77,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-6-flash","modelName":"Gemini 3.6 Flash","provider":"Google","rawName":"Gemini 3.6 Flash","variant":"epoch","setting":"Epoch AI run, accuracy","score":85.86,"unit":"%","rank":77,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-fable-5","modelName":"Claude Fable 5","provider":"Anthropic","rawName":"Claude Fable 5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":85.86,"unit":"%","rank":77,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 16k thinking","score":85.48,"unit":"%","rank":81,"rankComputedBySite":true,"date":"2025-11-25","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"nemotron-3-ultra","modelName":"Nemotron 3 Ultra","provider":"NVIDIA","rawName":"Nemotron 3 Ultra","variant":"epoch","setting":"Epoch AI run, accuracy","score":85.35,"unit":"%","rank":82,"rankComputedBySite":true,"date":"2026-08-10","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-8","modelName":"Claude Opus 4.8","provider":"Anthropic","rawName":"Claude Opus 4.8","variant":"epoch","setting":"Epoch AI run, accuracy","score":85.35,"unit":"%","rank":82,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":85.35,"unit":"%","rank":82,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"epoch","setting":"Epoch AI run, accuracy","score":85.29,"unit":"%","rank":85,"rankComputedBySite":true,"date":"2025-11-16","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-1","modelName":"GPT-5.1","provider":"OpenAI","rawName":"GPT-5.1","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":85.04,"unit":"%","rank":86,"rankComputedBySite":true,"date":"2025-11-17","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"kimi-k3","modelName":"Kimi K3","provider":"Moonshot AI","rawName":"Kimi K3","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":84.85,"unit":"%","rank":87,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen-3-5-plus-hosted-397b-a17b","modelName":"Qwen 3.5 Plus (hosted 397B-A17B)","provider":"Alibaba (Qwen)","rawName":"Qwen 3.5 Plus (hosted 397B-A17B)","variant":"epoch","setting":"Epoch AI run, accuracy","score":84.85,"unit":"%","rank":87,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4","modelName":"GPT-5.4","provider":"OpenAI","rawName":"GPT-5.4","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":84.85,"unit":"%","rank":87,"rankComputedBySite":true,"date":"2026-07-15","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-2-5-pro-jun-2025","modelName":"Gemini 2.5 Pro (Jun 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Jun 2025)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Jun 2025","score":84.85,"unit":"%","rank":87,"rankComputedBySite":true,"date":"2025-06-05","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"kimi-k2","modelName":"Kimi K2 Thinking","provider":"Moonshot AI","rawName":"Kimi K2 Thinking","variant":"epoch","setting":"Epoch AI run, accuracy","score":84.22,"unit":"%","rank":91,"rankComputedBySite":true,"date":"2025-11-11","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen-3-6-35b-a3b","modelName":"Qwen 3.6 35B-A3B","provider":"Alibaba (Qwen)","rawName":"Qwen 3.6 35B-A3B","variant":"epoch","setting":"Epoch AI run, accuracy","score":83.84,"unit":"%","rank":92,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-2-5-pro-mar-2025","modelName":"Gemini 2.5 Pro (Mar 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (Mar 2025)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Mar 2025","score":83.84,"unit":"%","rank":92,"rankComputedBySite":true,"date":"2025-03-31","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4-mini","modelName":"GPT-5.4 Mini","provider":"OpenAI","rawName":"GPT-5.4 Mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":83.59,"unit":"%","rank":94,"rankComputedBySite":true,"date":"2026-04-15","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v3-2","modelName":"DeepSeek-V3.2","provider":"DeepSeek","rawName":"DeepSeek-V3.2","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Thinking","score":83.42,"unit":"%","rank":95,"rankComputedBySite":true,"date":"2025-12-16","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen-3-6-flash","modelName":"Qwen 3.6 Flash","provider":"Alibaba (Qwen)","rawName":"Qwen 3.6 Flash","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 2026-04-16","score":83.33,"unit":"%","rank":96,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-fable-5","modelName":"Claude Fable 5","provider":"Anthropic","rawName":"Claude Fable 5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":83.33,"unit":"%","rank":96,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-6","modelName":"Claude Sonnet 4.6","provider":"Anthropic","rawName":"Claude Sonnet 4.6","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":83.33,"unit":"%","rank":96,"rankComputedBySite":true,"date":"2026-07-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-6","modelName":"Claude Sonnet 4.6","provider":"Anthropic","rawName":"Claude Sonnet 4.6","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":83.33,"unit":"%","rank":96,"rankComputedBySite":true,"date":"2026-07-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"glm-4-7","modelName":"GLM-4.7","provider":"Zhipu AI (Z.ai)","rawName":"GLM-4.7","variant":"epoch","setting":"Epoch AI run, accuracy","score":83.33,"unit":"%","rank":96,"rankComputedBySite":true,"date":"2026-01-29","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-6-sol","modelName":"GPT-5.6 Sol","provider":"OpenAI","rawName":"GPT-5.6 Sol","variant":"epoch","setting":"Epoch AI run, accuracy; effort: none","score":82.83,"unit":"%","rank":101,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":82.7,"unit":"%","rank":102,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-5-instant","modelName":"GPT-5.5 Instant","provider":"OpenAI","rawName":"GPT-5.5 Instant","variant":"epoch","setting":"Epoch AI run, accuracy","score":82.51,"unit":"%","rank":103,"rankComputedBySite":true,"date":"2026-08-02","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-6-luna","modelName":"GPT-5.6 Luna","provider":"OpenAI","rawName":"GPT-5.6 Luna","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":82.32,"unit":"%","rank":104,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-7-flash","modelName":"Qwen3.7 Flash","provider":"Alibaba (Qwen)","rawName":"Qwen3.7 Flash","variant":"epoch","setting":"Epoch AI run, accuracy","score":82.32,"unit":"%","rank":104,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen-3-5-flash-hosted-35b-a3b","modelName":"Qwen 3.5 Flash (hosted 35B-A3B)","provider":"Alibaba (Qwen)","rawName":"Qwen 3.5 Flash (hosted 35B-A3B)","variant":"epoch","setting":"Epoch AI run, accuracy","score":82.32,"unit":"%","rank":104,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 59k thinking","score":82.32,"unit":"%","rank":104,"rankComputedBySite":true,"date":"2025-10-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o3","modelName":"o3","provider":"OpenAI","rawName":"o3","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":81.82,"unit":"%","rank":108,"rankComputedBySite":true,"date":"2025-04-16","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 32k thinking","score":81.73,"unit":"%","rank":109,"rankComputedBySite":true,"date":"2025-10-21","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"minimax-m3","modelName":"MiniMax-M3","provider":"MiniMax","rawName":"MiniMax-M3","variant":"epoch","setting":"Epoch AI run, accuracy","score":81.31,"unit":"%","rank":110,"rankComputedBySite":true,"date":"2026-08-10","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-5-35b-a3b","modelName":"Qwen3.5-35B-A3B","provider":"Alibaba (Qwen)","rawName":"Qwen3.5-35B-A3B","variant":"epoch","setting":"Epoch AI run, accuracy","score":81.19,"unit":"%","rank":111,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o3","modelName":"o3","provider":"OpenAI","rawName":"o3","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":80.81,"unit":"%","rank":112,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-5","modelName":"Claude Opus 4.5","provider":"Anthropic","rawName":"Claude Opus 4.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: no thinking","score":80.68,"unit":"%","rank":113,"rankComputedBySite":true,"date":"2025-11-24","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-5","modelName":"Claude Sonnet 5","provider":"Anthropic","rawName":"Claude Sonnet 5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":80.3,"unit":"%","rank":114,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-235b-a22b-thinking-jul-2025","modelName":"Qwen3-235B-A22B-Thinking (Jul 2025)","provider":"Alibaba (Qwen)","rawName":"Qwen3-235B-A22B-Thinking (Jul 2025)","variant":"epoch","setting":"Epoch AI run, accuracy","score":80.05,"unit":"%","rank":115,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o3","modelName":"o3","provider":"OpenAI","rawName":"o3","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":79.8,"unit":"%","rank":116,"rankComputedBySite":true,"date":"2026-07-15","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-7-sonnet","modelName":"Claude 3.7 Sonnet","provider":"Anthropic","rawName":"Claude 3.7 Sonnet","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 64k thinking","score":79.73,"unit":"%","rank":117,"rankComputedBySite":true,"date":"2025-05-26","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o4-mini","modelName":"o4-mini","provider":"OpenAI","rawName":"o4-mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":79.61,"unit":"%","rank":118,"rankComputedBySite":true,"date":"2025-04-16","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4","modelName":"Claude Sonnet 4","provider":"Anthropic","rawName":"Claude Sonnet 4","variant":"epoch","setting":"Epoch AI run, accuracy","score":79.19,"unit":"%","rank":119,"rankComputedBySite":true,"date":"2025-05-26","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-5-9b","modelName":"Qwen3.5-9B","provider":"Alibaba (Qwen)","rawName":"Qwen3.5-9B","variant":"epoch","setting":"Epoch AI run, accuracy","score":78.91,"unit":"%","rank":120,"rankComputedBySite":true,"date":"2026-08-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-6","modelName":"Claude Sonnet 4.6","provider":"Anthropic","rawName":"Claude Sonnet 4.6","variant":"epoch","setting":"Epoch AI run, accuracy; effort: max","score":78.79,"unit":"%","rank":121,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-fable-5","modelName":"Claude Fable 5","provider":"Anthropic","rawName":"Claude Fable 5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":78.79,"unit":"%","rank":121,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 16k thinking","score":78.79,"unit":"%","rank":121,"rankComputedBySite":true,"date":"2025-10-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4-nano","modelName":"GPT-5.4 Nano","provider":"OpenAI","rawName":"GPT-5.4 Nano","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":78.47,"unit":"%","rank":124,"rankComputedBySite":true,"date":"2026-04-14","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o4-mini","modelName":"o4-mini","provider":"OpenAI","rawName":"o4-mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":77.78,"unit":"%","rank":125,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-5","modelName":"GPT-5.5","provider":"OpenAI","rawName":"GPT-5.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: no thinking","score":77.27,"unit":"%","rank":126,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-6-terra","modelName":"GPT-5.6 Terra","provider":"OpenAI","rawName":"GPT-5.6 Terra","variant":"epoch","setting":"Epoch AI run, accuracy; effort: none","score":77.27,"unit":"%","rank":126,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o3-mini","modelName":"o3-mini","provider":"OpenAI","rawName":"o3-mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":77.02,"unit":"%","rank":128,"rankComputedBySite":true,"date":"2025-02-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4-1","modelName":"Claude Opus 4.1","provider":"Anthropic","rawName":"Claude Opus 4.1","variant":"epoch","setting":"Epoch AI run, accuracy","score":76.77,"unit":"%","rank":129,"rankComputedBySite":true,"date":"2025-08-05","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-7-sonnet","modelName":"Claude 3.7 Sonnet","provider":"Anthropic","rawName":"Claude 3.7 Sonnet","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 32k thinking","score":76.77,"unit":"%","rank":129,"rankComputedBySite":true,"date":"2025-03-10","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-7-sonnet","modelName":"Claude 3.7 Sonnet","provider":"Anthropic","rawName":"Claude 3.7 Sonnet","variant":"epoch","setting":"Epoch AI run, accuracy; effort: 16k thinking","score":76.77,"unit":"%","rank":129,"rankComputedBySite":true,"date":"2025-02-26","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o1","modelName":"o1","provider":"OpenAI","rawName":"o1","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":76.77,"unit":"%","rank":129,"rankComputedBySite":true,"date":"2025-02-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-r1-may-2025","modelName":"DeepSeek-R1 (May 2025)","provider":"DeepSeek","rawName":"DeepSeek-R1 (May 2025)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: May 2025","score":76.33,"unit":"%","rank":133,"rankComputedBySite":true,"date":"2025-05-29","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-opus-4","modelName":"Claude Opus 4","provider":"Anthropic","rawName":"Claude Opus 4","variant":"epoch","setting":"Epoch AI run, accuracy","score":76.26,"unit":"%","rank":134,"rankComputedBySite":true,"date":"2025-05-22","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemma-4-31b-it","modelName":"Gemma 4 31B IT","provider":"Google","rawName":"Gemma 4 31B IT","variant":"epoch","setting":"Epoch AI run, accuracy","score":75.76,"unit":"%","rank":135,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-5-flash-lite","modelName":"Gemini 3.5 Flash-Lite","provider":"Google","rawName":"Gemini 3.5 Flash-Lite","variant":"epoch","setting":"Epoch AI run, accuracy","score":75.76,"unit":"%","rank":135,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-oss-120b","modelName":"gpt-oss-120b","provider":"OpenAI","rawName":"gpt-oss-120b","variant":"epoch","setting":"Epoch AI run, accuracy","score":75.76,"unit":"%","rank":135,"rankComputedBySite":true,"date":"2025-12-11","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"grok-3","modelName":"Grok 3","provider":"xAI","rawName":"Grok 3","variant":"epoch","setting":"Epoch AI run, accuracy","score":75.76,"unit":"%","rank":135,"rankComputedBySite":true,"date":"2025-05-26","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o1","modelName":"o1","provider":"OpenAI","rawName":"o1","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":75.76,"unit":"%","rank":135,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"grok-3-mini","modelName":"Grok-3 mini","provider":"xAI","rawName":"Grok-3 mini","variant":"epoch","setting":"Epoch AI run, accuracy","score":75.51,"unit":"%","rank":140,"rankComputedBySite":true,"date":"2025-05-26","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o4-mini","modelName":"o4-mini","provider":"OpenAI","rawName":"o4-mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":75.25,"unit":"%","rank":141,"rankComputedBySite":true,"date":"2026-07-11","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":75,"unit":"%","rank":142,"rankComputedBySite":true,"date":"2025-10-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4","modelName":"GPT-5.4","provider":"OpenAI","rawName":"GPT-5.4","variant":"epoch","setting":"Epoch AI run, accuracy; effort: none","score":74.75,"unit":"%","rank":143,"rankComputedBySite":true,"date":"2026-07-15","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o3-mini","modelName":"o3-mini","provider":"OpenAI","rawName":"o3-mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":74.27,"unit":"%","rank":144,"rankComputedBySite":true,"date":"2025-01-31","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o1","modelName":"o1","provider":"OpenAI","rawName":"o1","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":74.24,"unit":"%","rank":145,"rankComputedBySite":true,"date":"2026-07-20","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-3-1-flash-lite","modelName":"Gemini 3.1 Flash-Lite","provider":"Google","rawName":"Gemini 3.1 Flash-Lite","variant":"epoch","setting":"Epoch AI run, accuracy","score":73.74,"unit":"%","rank":146,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-sonnet-4-5","modelName":"Claude Sonnet 4.5","provider":"Anthropic","rawName":"Claude Sonnet 4.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: no thinking","score":73.74,"unit":"%","rank":146,"rankComputedBySite":true,"date":"2025-09-29","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemma-4-26b-a4b","modelName":"Gemma 4 26B A4B","provider":"Google","rawName":"Gemma 4 26B A4B","variant":"epoch","setting":"Epoch AI run, accuracy","score":73.23,"unit":"%","rank":148,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v4-pro","modelName":"DeepSeek-V4-Pro","provider":"DeepSeek","rawName":"DeepSeek-V4-Pro","variant":"epoch","setting":"Epoch AI run, accuracy; effort: unknown thinking","score":73.23,"unit":"%","rank":148,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-2","modelName":"GPT-5.2","provider":"OpenAI","rawName":"GPT-5.2","variant":"epoch","setting":"Epoch AI run, accuracy; effort: none","score":73.23,"unit":"%","rank":148,"rankComputedBySite":true,"date":"2026-07-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-max","modelName":"Qwen3-Max","provider":"Alibaba (Qwen)","rawName":"Qwen3-Max","variant":"epoch","setting":"Epoch AI run, accuracy","score":72.6,"unit":"%","rank":151,"rankComputedBySite":true,"date":"2025-10-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4-nano","modelName":"GPT-5.4 Nano","provider":"OpenAI","rawName":"GPT-5.4 Nano","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":72.22,"unit":"%","rank":152,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: minimal","score":71.72,"unit":"%","rank":153,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5","modelName":"GPT-5","provider":"OpenAI","rawName":"GPT-5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: minimal","score":71.72,"unit":"%","rank":153,"rankComputedBySite":true,"date":"2026-07-20","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-r1","modelName":"DeepSeek-R1","provider":"DeepSeek","rawName":"DeepSeek-R1","variant":"epoch","setting":"Epoch AI run, accuracy","score":71.72,"unit":"%","rank":153,"rankComputedBySite":true,"date":"2025-05-26","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-mini","modelName":"GPT-5 mini","provider":"OpenAI","rawName":"GPT-5 mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":71.65,"unit":"%","rank":156,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"seed-oss-36b-instruct","modelName":"seed-oss-36b-instruct","provider":"ByteDance","rawName":"seed-oss-36b-instruct","variant":"epoch","setting":"Epoch AI run, accuracy","score":71.53,"unit":"%","rank":157,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"glm-5-2","modelName":"GLM-5.2","provider":"Zhipu AI (Z.ai)","rawName":"GLM-5.2","variant":"epoch","setting":"Epoch AI run, accuracy","score":71.21,"unit":"%","rank":158,"rankComputedBySite":true,"date":"2026-08-10","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v3-2","modelName":"DeepSeek-V3.2","provider":"DeepSeek","rawName":"DeepSeek-V3.2","variant":"epoch","setting":"Epoch AI run, accuracy","score":71.21,"unit":"%","rank":158,"rankComputedBySite":true,"date":"2026-07-16","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-haiku-4-5","modelName":"Claude Haiku 4.5","provider":"Anthropic","rawName":"Claude Haiku 4.5","variant":"epoch","setting":"Epoch AI run, accuracy","score":71.21,"unit":"%","rank":158,"rankComputedBySite":true,"date":"2025-10-22","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-235b-a22b","modelName":"Qwen3-235B-A22B","provider":"Alibaba (Qwen)","rawName":"Qwen3-235B-A22B","variant":"epoch","setting":"Epoch AI run, accuracy","score":70.71,"unit":"%","rank":161,"rankComputedBySite":true,"date":"2025-06-03","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-30b-a3b-thinking-jul-2025","modelName":"Qwen3-30B-A3B-Thinking (Jul 2025)","provider":"Alibaba (Qwen)","rawName":"Qwen3-30B-A3B-Thinking (Jul 2025)","variant":"epoch","setting":"Epoch AI run, accuracy","score":70.08,"unit":"%","rank":162,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":69.44,"unit":"%","rank":163,"rankComputedBySite":true,"date":"2025-10-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4-5","modelName":"GPT-4.5","provider":"OpenAI","rawName":"GPT-4.5","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Feb 2025","score":68.69,"unit":"%","rank":164,"rankComputedBySite":true,"date":"2025-02-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o3-mini","modelName":"o3-mini","provider":"OpenAI","rawName":"o3-mini","variant":"epoch","setting":"Epoch AI run, accuracy","score":68.18,"unit":"%","rank":165,"rankComputedBySite":true,"date":"2026-07-20","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v3-mar-2025","modelName":"DeepSeek-V3 (Mar 2025)","provider":"DeepSeek","rawName":"DeepSeek-V3 (Mar 2025)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Mar 2025","score":67.61,"unit":"%","rank":166,"rankComputedBySite":true,"date":"2025-04-01","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":67.42,"unit":"%","rank":167,"rankComputedBySite":true,"date":"2025-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-4-maverick","modelName":"Llama 4 Maverick","provider":"Meta","rawName":"Llama 4 Maverick","variant":"epoch","setting":"Epoch AI run, accuracy; effort: FP8","score":66.98,"unit":"%","rank":168,"rankComputedBySite":true,"date":"2025-04-08","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4-1","modelName":"GPT-4.1","provider":"OpenAI","rawName":"GPT-4.1","variant":"epoch","setting":"Epoch AI run, accuracy","score":66.92,"unit":"%","rank":169,"rankComputedBySite":true,"date":"2025-04-14","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-1","modelName":"GPT-5.1","provider":"OpenAI","rawName":"GPT-5.1","variant":"epoch","setting":"Epoch AI run, accuracy; effort: no thinking","score":66.67,"unit":"%","rank":170,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-2-5-pro-may-2025","modelName":"Gemini 2.5 Pro (May 2025)","provider":"Google","rawName":"Gemini 2.5 Pro (May 2025)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Jun 2025","score":66.67,"unit":"%","rank":170,"rankComputedBySite":true,"date":"2025-06-03","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-7-sonnet","modelName":"Claude 3.7 Sonnet","provider":"Anthropic","rawName":"Claude 3.7 Sonnet","variant":"epoch","setting":"Epoch AI run, accuracy","score":66.04,"unit":"%","rank":172,"rankComputedBySite":true,"date":"2025-02-24","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4-1-mini","modelName":"GPT-4.1 mini","provider":"OpenAI","rawName":"GPT-4.1 mini","variant":"epoch","setting":"Epoch AI run, accuracy","score":65.85,"unit":"%","rank":173,"rankComputedBySite":true,"date":"2025-04-14","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-2-0-pro","modelName":"Gemini 2.0 Pro","provider":"Google","rawName":"Gemini 2.0 Pro","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Feb 2025","score":65.66,"unit":"%","rank":174,"rankComputedBySite":true,"date":"2025-02-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwq-plus","modelName":"QWQ-Plus","provider":"Alibaba (Qwen)","rawName":"QWQ-Plus","variant":"epoch","setting":"Epoch AI run, accuracy","score":65.4,"unit":"%","rank":175,"rankComputedBySite":true,"date":"2025-04-11","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwq-32b","modelName":"QwQ-32B","provider":"Alibaba (Qwen)","rawName":"QwQ-32B","variant":"epoch","setting":"Epoch AI run, accuracy","score":65.34,"unit":"%","rank":176,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-r1-distill-qwen-32b","modelName":"DeepSeek-R1-Distill-Qwen-32B","provider":"DeepSeek","rawName":"DeepSeek-R1-Distill-Qwen-32B","variant":"epoch","setting":"Epoch AI run, accuracy","score":64.14,"unit":"%","rank":177,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4-mini","modelName":"GPT-5.4 Mini","provider":"OpenAI","rawName":"GPT-5.4 Mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: none","score":64.14,"unit":"%","rank":177,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-2-0-flash-feb-2025","modelName":"Gemini 2.0 Flash (Feb 2025)","provider":"Google","rawName":"Gemini 2.0 Flash (Feb 2025)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Feb 2025","score":64.14,"unit":"%","rank":177,"rankComputedBySite":true,"date":"2025-02-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-6-luna","modelName":"GPT-5.6 Luna","provider":"OpenAI","rawName":"GPT-5.6 Luna","variant":"epoch","setting":"Epoch AI run, accuracy; effort: none","score":63.64,"unit":"%","rank":180,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o1-mini","modelName":"o1-mini","provider":"OpenAI","rawName":"o1-mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: high","score":62.37,"unit":"%","rank":181,"rankComputedBySite":true,"date":"2025-02-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-oss-20b","modelName":"gpt-oss-20b","provider":"OpenAI","rawName":"gpt-oss-20b","variant":"epoch","setting":"Epoch AI run, accuracy","score":60.8,"unit":"%","rank":182,"rankComputedBySite":true,"date":"2026-08-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-medium-3","modelName":"Mistral Medium 3","provider":"Mistral AI","rawName":"Mistral Medium 3","variant":"epoch","setting":"Epoch AI run, accuracy","score":59.53,"unit":"%","rank":183,"rankComputedBySite":true,"date":"2025-05-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o1-mini","modelName":"o1-mini","provider":"OpenAI","rawName":"o1-mini","variant":"epoch","setting":"Epoch AI run, accuracy; effort: medium","score":59.5,"unit":"%","rank":184,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"epoch","setting":"Epoch AI run, accuracy; effort: low","score":57.58,"unit":"%","rank":185,"rankComputedBySite":true,"date":"2026-07-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-1-5-pro-sept-2024","modelName":"Gemini 1.5 Pro (Sept 2024)","provider":"Google","rawName":"Gemini 1.5 Pro (Sept 2024)","variant":"epoch","setting":"Epoch AI run, accuracy","score":57.23,"unit":"%","rank":186,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-2-0-flash-thinking-jan-2025","modelName":"Gemini 2.0 Flash Thinking (Jan 2025)","provider":"Google","rawName":"Gemini 2.0 Flash Thinking (Jan 2025)","variant":"epoch","setting":"Epoch AI run, accuracy","score":57.07,"unit":"%","rank":187,"rankComputedBySite":true,"date":"2025-02-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-v3","modelName":"DeepSeek-V3","provider":"DeepSeek","rawName":"DeepSeek-V3","variant":"epoch","setting":"Epoch AI run, accuracy","score":56.53,"unit":"%","rank":188,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen2-5-max","modelName":"Qwen2.5-Max","provider":"Alibaba (Qwen)","rawName":"Qwen2.5-Max","variant":"epoch","setting":"Epoch AI run, accuracy","score":56.12,"unit":"%","rank":189,"rankComputedBySite":true,"date":"2025-04-01","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"magistral-small-1-0","modelName":"Magistral Small 1.0","provider":"Mistral AI","rawName":"Magistral Small 1.0","variant":"epoch","setting":"Epoch AI run, accuracy","score":56.06,"unit":"%","rank":190,"rankComputedBySite":true,"date":"2026-08-06","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"phi-4","modelName":"Phi-4","provider":"Microsoft Research","rawName":"Phi-4","variant":"epoch","setting":"Epoch AI run, accuracy","score":56.06,"unit":"%","rank":190,"rankComputedBySite":true,"date":"2025-01-31","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-r1-distill-llama-70b","modelName":"DeepSeek-R1-Distill-Llama-70B","provider":"DeepSeek","rawName":"DeepSeek-R1-Distill-Llama-70B","variant":"epoch","setting":"Epoch AI run, accuracy","score":55.74,"unit":"%","rank":192,"rankComputedBySite":true,"date":"2025-03-10","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-30b-a3b-instruct-jul-2025","modelName":"Qwen3-30B-A3B-Instruct (Jul 2025)","provider":"Alibaba (Qwen)","rawName":"Qwen3-30B-A3B-Instruct (Jul 2025)","variant":"epoch","setting":"Epoch AI run, accuracy","score":55.62,"unit":"%","rank":193,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-4-nano","modelName":"GPT-5.4 Nano","provider":"OpenAI","rawName":"GPT-5.4 Nano","variant":"epoch","setting":"Epoch AI run, accuracy; effort: no thinking","score":55.56,"unit":"%","rank":194,"rankComputedBySite":true,"date":"2026-08-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-5-sonnet-october-2024","modelName":"Claude 3.5 Sonnet (October 2024)","provider":"Anthropic","rawName":"Claude 3.5 Sonnet (October 2024)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Oct 2024","score":55.3,"unit":"%","rank":195,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-32b","modelName":"Qwen3-32B","provider":"Alibaba (Qwen)","rawName":"Qwen3-32B","variant":"epoch","setting":"Epoch AI run, accuracy","score":54.1,"unit":"%","rank":196,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-5-sonnet","modelName":"Claude 3.5 Sonnet","provider":"Anthropic","rawName":"Claude 3.5 Sonnet","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Jun 2024","score":54.04,"unit":"%","rank":197,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"grok-2-dec-2024","modelName":"Grok-2 (Dec 2024)","provider":"xAI","rawName":"Grok-2 (Dec 2024)","variant":"epoch","setting":"Epoch AI run, accuracy","score":53.79,"unit":"%","rank":198,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-14b","modelName":"Qwen3-14B","provider":"Alibaba (Qwen)","rawName":"Qwen3-14B","variant":"epoch","setting":"Epoch AI run, accuracy","score":53.41,"unit":"%","rank":199,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-4-scout","modelName":"Llama 4 Scout","provider":"Meta","rawName":"Llama 4 Scout","variant":"epoch","setting":"Epoch AI run, accuracy","score":51.83,"unit":"%","rank":200,"rankComputedBySite":true,"date":"2025-04-08","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-large-2-nov-2024","modelName":"Mistral Large 2 (Nov 2024)","provider":"Mistral AI","rawName":"Mistral Large 2 (Nov 2024)","variant":"epoch","setting":"Epoch AI run, accuracy","score":51.33,"unit":"%","rank":201,"rankComputedBySite":true,"date":"2025-02-25","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-3-1-405b","modelName":"Llama 3.1-405B","provider":"Meta","rawName":"Llama 3.1-405B","variant":"epoch","setting":"Epoch AI run, accuracy","score":50.92,"unit":"%","rank":202,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-30b-a3b","modelName":"Qwen3-30B-A3B","provider":"Alibaba (Qwen)","rawName":"Qwen3-30B-A3B","variant":"epoch","setting":"Epoch AI run, accuracy","score":50.38,"unit":"%","rank":203,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"o1-preview","modelName":"o1-preview","provider":"OpenAI","rawName":"o1-preview","variant":"epoch","setting":"Epoch AI run, accuracy","score":50.32,"unit":"%","rank":204,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4o-aug-2024","modelName":"GPT-4o (Aug 2024)","provider":"OpenAI","rawName":"GPT-4o (Aug 2024)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Aug 2024","score":49.21,"unit":"%","rank":205,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen2-5-72b","modelName":"Qwen2.5-72B","provider":"Alibaba (Qwen)","rawName":"Qwen2.5-72B","variant":"epoch","setting":"Epoch AI run, accuracy","score":49.15,"unit":"%","rank":206,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-small-3-2","modelName":"Mistral Small 3.2","provider":"Mistral AI","rawName":"Mistral Small 3.2","variant":"epoch","setting":"Epoch AI run, accuracy","score":49.05,"unit":"%","rank":207,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-large-2-jul-2024","modelName":"Mistral Large 2 (Jul 2024)","provider":"Mistral AI","rawName":"Mistral Large 2 (Jul 2024)","variant":"epoch","setting":"Epoch AI run, accuracy","score":49.02,"unit":"%","rank":208,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4-1-nano","modelName":"GPT-4.1 nano","provider":"OpenAI","rawName":"GPT-4.1 nano","variant":"epoch","setting":"Epoch AI run, accuracy","score":48.93,"unit":"%","rank":209,"rankComputedBySite":true,"date":"2025-04-14","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4o-may-2024","modelName":"GPT-4o (May 2024)","provider":"OpenAI","rawName":"GPT-4o (May 2024)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: May 2024","score":48.9,"unit":"%","rank":210,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-5-nano","modelName":"GPT-5 nano","provider":"OpenAI","rawName":"GPT-5 nano","variant":"epoch","setting":"Epoch AI run, accuracy; effort: minimal","score":48.48,"unit":"%","rank":211,"rankComputedBySite":true,"date":"2026-07-13","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen-plus-jan-2025","modelName":"Qwen Plus (Jan 2025)","provider":"Alibaba (Qwen)","rawName":"Qwen Plus (Jan 2025)","variant":"epoch","setting":"Epoch AI run, accuracy","score":48.11,"unit":"%","rank":212,"rankComputedBySite":true,"date":"2025-04-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4o-nov-2024","modelName":"GPT-4o (Nov 2024)","provider":"OpenAI","rawName":"GPT-4o (Nov 2024)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Nov 2024","score":47.89,"unit":"%","rank":213,"rankComputedBySite":true,"date":"2025-02-05","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemma-3-27b","modelName":"Gemma 3 27B","provider":"Google","rawName":"Gemma 3 27B","variant":"epoch","setting":"Epoch AI run, accuracy","score":47.73,"unit":"%","rank":214,"rankComputedBySite":true,"date":"2026-08-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"magistral-small-1-2","modelName":"Magistral Small 1.2","provider":"Mistral AI","rawName":"Magistral Small 1.2","variant":"epoch","setting":"Epoch AI run, accuracy","score":47.6,"unit":"%","rank":215,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-8b","modelName":"Qwen3-8B","provider":"Alibaba (Qwen)","rawName":"Qwen3-8B","variant":"epoch","setting":"Epoch AI run, accuracy","score":47.47,"unit":"%","rank":216,"rankComputedBySite":true,"date":"2026-08-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-3-3-70b","modelName":"Llama 3.3 70B","provider":"Meta","rawName":"Llama 3.3 70B","variant":"epoch","setting":"Epoch AI run, accuracy","score":47.44,"unit":"%","rank":217,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-1-5-flash-sep-2024","modelName":"Gemini 1.5 Flash (Sep 2024)","provider":"Google","rawName":"Gemini 1.5 Flash (Sep 2024)","variant":"epoch","setting":"Epoch AI run, accuracy","score":47.32,"unit":"%","rank":218,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-small-3","modelName":"Mistral Small 3","provider":"Mistral AI","rawName":"Mistral Small 3","variant":"epoch","setting":"Epoch AI run, accuracy","score":47.29,"unit":"%","rank":219,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-opus","modelName":"Claude 3 Opus","provider":"Anthropic","rawName":"Claude 3 Opus","variant":"epoch","setting":"Epoch AI run, accuracy","score":47.16,"unit":"%","rank":220,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4-turbo-apr-2024","modelName":"GPT-4 Turbo (Apr 2024)","provider":"OpenAI","rawName":"GPT-4 Turbo (Apr 2024)","variant":"epoch","setting":"Epoch AI run, accuracy","score":46.59,"unit":"%","rank":221,"rankComputedBySite":true,"date":"2025-02-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"tulu-3-t-lu-3-70b","modelName":"Tulu 3 (Tülu 3) 70B","provider":"Ai2","rawName":"Tulu 3 (Tülu 3) 70B","variant":"epoch","setting":"Epoch AI run, accuracy","score":46.28,"unit":"%","rank":222,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen2-5-32b","modelName":"Qwen2.5-32B","provider":"Alibaba (Qwen)","rawName":"Qwen2.5-32B","variant":"epoch","setting":"Epoch AI run, accuracy","score":46.09,"unit":"%","rank":223,"rankComputedBySite":true,"date":"2025-01-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-1-5-pro-may-2024","modelName":"Gemini 1.5 Pro (May 2024)","provider":"Google","rawName":"Gemini 1.5 Pro (May 2024)","variant":"epoch","setting":"Epoch AI run, accuracy","score":45.86,"unit":"%","rank":224,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-4b-instruct-2507","modelName":"qwen3-4b-instruct-2507","provider":"Alibaba (Qwen)","rawName":"qwen3-4b-instruct-2507","variant":"epoch","setting":"Epoch AI run, accuracy","score":45.83,"unit":"%","rank":225,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"glm-4-7-flash","modelName":"glm-4.7-flash_none","provider":"Zhipu AI (Z.ai)","rawName":"glm-4.7-flash_none","variant":"epoch","setting":"Epoch AI run, accuracy","score":45.14,"unit":"%","rank":226,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-r1-distill-qwen-14b","modelName":"DeepSeek-R1-Distill-Qwen-14B","provider":"DeepSeek","rawName":"DeepSeek-R1-Distill-Qwen-14B","variant":"epoch","setting":"Epoch AI run, accuracy","score":44.7,"unit":"%","rank":227,"rankComputedBySite":true,"date":"2025-03-10","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-3-1-70b","modelName":"Llama 3.1-70B","provider":"Meta","rawName":"Llama 3.1-70B","variant":"epoch","setting":"Epoch AI run, accuracy","score":44.19,"unit":"%","rank":228,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-4b","modelName":"Qwen3-4B","provider":"Alibaba (Qwen)","rawName":"Qwen3-4B","variant":"epoch","setting":"Epoch AI run, accuracy","score":43.56,"unit":"%","rank":229,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"wizardlm-2-8x22b","modelName":"WizardLM-2 8x22B","provider":"Microsoft","rawName":"WizardLM-2 8x22B","variant":"epoch","setting":"Epoch AI run, accuracy","score":43.43,"unit":"%","rank":230,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4-turbo-nov-2023","modelName":"GPT-4 Turbo (Nov 2023)","provider":"OpenAI","rawName":"GPT-4 Turbo (Nov 2023)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Nov 2023","score":42.36,"unit":"%","rank":231,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4-turbo-nov-2023","modelName":"GPT-4 Turbo (Nov 2023)","provider":"OpenAI","rawName":"GPT-4 Turbo (Nov 2023)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: January 2024","score":42.27,"unit":"%","rank":232,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-small-3-1","modelName":"Mistral Small 3.1","provider":"Mistral AI","rawName":"Mistral Small 3.1","variant":"epoch","setting":"Epoch AI run, accuracy","score":41.92,"unit":"%","rank":233,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen-turbo","modelName":"Qwen-Turbo","provider":"Alibaba (Qwen)","rawName":"Qwen-Turbo","variant":"epoch","setting":"Epoch AI run, accuracy","score":41.79,"unit":"%","rank":234,"rankComputedBySite":true,"date":"2025-04-07","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-3-2-90b","modelName":"Llama 3.2 90B","provider":"Meta","rawName":"Llama 3.2 90B","variant":"epoch","setting":"Epoch AI run, accuracy","score":41.04,"unit":"%","rank":235,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen2-72b","modelName":"Qwen2-72B","provider":"Alibaba (Qwen)","rawName":"Qwen2-72B","variant":"epoch","setting":"Epoch AI run, accuracy","score":40.78,"unit":"%","rank":236,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-sonnet","modelName":"Claude 3 Sonnet","provider":"Anthropic","rawName":"Claude 3 Sonnet","variant":"epoch","setting":"Epoch AI run, accuracy","score":40.59,"unit":"%","rank":237,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-3-70b","modelName":"Llama 3-70B","provider":"Meta","rawName":"Llama 3-70B","variant":"epoch","setting":"Epoch AI run, accuracy","score":40.56,"unit":"%","rank":238,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-1-5-flash-may-2024","modelName":"Gemini 1.5 Flash (May 2024)","provider":"Google","rawName":"Gemini 1.5 Flash (May 2024)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: May 2024","score":40.37,"unit":"%","rank":239,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemma-3-12b","modelName":"Gemma 3 12B","provider":"Google","rawName":"Gemma 3 12B","variant":"epoch","setting":"Epoch AI run, accuracy","score":39.46,"unit":"%","rank":240,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-large","modelName":"Mistral Large","provider":"Mistral AI","rawName":"Mistral Large","variant":"epoch","setting":"Epoch AI run, accuracy","score":38.76,"unit":"%","rank":241,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-5-haiku","modelName":"Claude 3.5 Haiku","provider":"Anthropic","rawName":"Claude 3.5 Haiku","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Oct 2024","score":38.13,"unit":"%","rank":242,"rankComputedBySite":true,"date":"2025-03-12","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4o-mini","modelName":"GPT-4o mini","provider":"OpenAI","rawName":"GPT-4o mini","variant":"epoch","setting":"Epoch AI run, accuracy","score":37.72,"unit":"%","rank":243,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"hermes-2-theta-llama-3-70b","modelName":"Hermes 2 Theta Llama-3 70B","provider":"Nous Research","rawName":"Hermes 2 Theta Llama-3 70B","variant":"epoch","setting":"Epoch AI run, accuracy","score":37.47,"unit":"%","rank":244,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemma-2-27b","modelName":"Gemma 2 27B","provider":"Google","rawName":"Gemma 2 27B","variant":"epoch","setting":"Epoch AI run, accuracy","score":36.49,"unit":"%","rank":245,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-3-haiku","modelName":"Claude 3 Haiku","provider":"Anthropic","rawName":"Claude 3 Haiku","variant":"epoch","setting":"Epoch AI run, accuracy","score":36.3,"unit":"%","rank":246,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4-mar-2023","modelName":"GPT-4 (Mar 2023)","provider":"OpenAI","rawName":"GPT-4 (Mar 2023)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Mar 2023","score":35.73,"unit":"%","rank":247,"rankComputedBySite":true,"date":"2025-10-23","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen2-5-7b","modelName":"Qwen2.5-7B","provider":"Alibaba (Qwen)","rawName":"Qwen2.5-7B","variant":"epoch","setting":"Epoch AI run, accuracy","score":35.48,"unit":"%","rank":248,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-2","modelName":"Claude 2","provider":"Anthropic","rawName":"Claude 2","variant":"epoch","setting":"Epoch AI run, accuracy","score":34.66,"unit":"%","rank":249,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mixtral-8x22b","modelName":"Mixtral 8x22B","provider":"Mistral AI","rawName":"Mixtral 8x22B","variant":"epoch","setting":"Epoch AI run, accuracy","score":34.06,"unit":"%","rank":250,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-1-0-pro","modelName":"Gemini 1.0 Pro","provider":"Google","rawName":"Gemini 1.0 Pro","variant":"epoch","setting":"Epoch AI run, accuracy","score":33.96,"unit":"%","rank":251,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"eurus-2-7b-prime","modelName":"Eurus-2-7B-PRIME","provider":"Tsinghua University","rawName":"Eurus-2-7B-PRIME","variant":"epoch","setting":"Epoch AI run, accuracy","score":33.9,"unit":"%","rank":252,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-r1-distill-qwen-1-5b","modelName":"DeepSeek-R1-Distill-Qwen-1.5B","provider":"DeepSeek","rawName":"DeepSeek-R1-Distill-Qwen-1.5B","variant":"epoch","setting":"Epoch AI run, accuracy","score":33.59,"unit":"%","rank":253,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemini-1-5-flash-8b","modelName":"Gemini 1.5 Flash 8B","provider":"Google","rawName":"Gemini 1.5 Flash 8B","variant":"epoch","setting":"Epoch AI run, accuracy","score":32.95,"unit":"%","rank":254,"rankComputedBySite":true,"date":"2025-03-12","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"claude-2-1","modelName":"Claude 2.1","provider":"Anthropic","rawName":"Claude 2.1","variant":"epoch","setting":"Epoch AI run, accuracy","score":32.95,"unit":"%","rank":254,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"dbrx","modelName":"DBRX","provider":"Databricks","rawName":"DBRX","variant":"epoch","setting":"Epoch AI run, accuracy","score":32.89,"unit":"%","rank":256,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"yi-1-5-34b","modelName":"Yi-1.5-34B","provider":"01.AI","rawName":"Yi-1.5-34B","variant":"epoch","setting":"Epoch AI run, accuracy","score":31.98,"unit":"%","rank":257,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen1-5-32b","modelName":"Qwen1.5-32B","provider":"Alibaba (Qwen)","rawName":"Qwen1.5-32B","variant":"epoch","setting":"Epoch AI run, accuracy","score":30.74,"unit":"%","rank":258,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-4-jun-2023","modelName":"GPT-4 (Jun 2023)","provider":"OpenAI","rawName":"GPT-4 (Jun 2023)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Jun 2023","score":30.65,"unit":"%","rank":259,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen3-1-7b","modelName":"Qwen3-1.7B","provider":"Alibaba (Qwen)","rawName":"Qwen3-1.7B","variant":"epoch","setting":"Epoch AI run, accuracy","score":30.62,"unit":"%","rank":260,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mixtral-8x7b","modelName":"Mixtral 8x7B","provider":"Mistral AI","rawName":"Mixtral 8x7B","variant":"epoch","setting":"Epoch AI run, accuracy","score":30.59,"unit":"%","rank":261,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-nemo","modelName":"Mistral NeMo","provider":"Mistral AI","rawName":"Mistral NeMo","variant":"epoch","setting":"Epoch AI run, accuracy","score":29.89,"unit":"%","rank":262,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"qwen1-5-72b","modelName":"Qwen1.5-72B","provider":"Alibaba (Qwen)","rawName":"Qwen1.5-72B","variant":"epoch","setting":"Epoch AI run, accuracy","score":28.82,"unit":"%","rank":263,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"granite-4-0-micro","modelName":"granite-4.0-micro","provider":"IBM","rawName":"granite-4.0-micro","variant":"epoch","setting":"Epoch AI run, accuracy","score":28.28,"unit":"%","rank":264,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-3-5-turbo-nov-2023","modelName":"GPT-3.5 Turbo (Nov 2023)","provider":"OpenAI","rawName":"GPT-3.5 Turbo (Nov 2023)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Nov 2023","score":28.03,"unit":"%","rank":265,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"phi-3-medium-14b","modelName":"phi-3-medium 14B","provider":"Microsoft","rawName":"phi-3-medium 14B","variant":"epoch","setting":"Epoch AI run, accuracy","score":27.59,"unit":"%","rank":266,"rankComputedBySite":true,"date":"2025-01-31","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemma-2-9b","modelName":"Gemma 2 9B","provider":"Google","rawName":"Gemma 2 9B","variant":"epoch","setting":"Epoch AI run, accuracy","score":27.46,"unit":"%","rank":267,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gpt-3-5-turbo-jan-2024","modelName":"GPT-3.5 Turbo (Jan 2024)","provider":"OpenAI","rawName":"GPT-3.5 Turbo (Jan 2024)","variant":"epoch","setting":"Epoch AI run, accuracy; effort: Jan 2024","score":27.18,"unit":"%","rank":268,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"ministral-8b","modelName":"Ministral 8B","provider":"Mistral AI","rawName":"Ministral 8B","variant":"epoch","setting":"Epoch AI run, accuracy","score":27.15,"unit":"%","rank":269,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-3-1-8b","modelName":"Llama 3.1-8B","provider":"Meta","rawName":"Llama 3.1-8B","variant":"epoch","setting":"Epoch AI run, accuracy","score":26.96,"unit":"%","rank":270,"rankComputedBySite":true,"date":"2026-08-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-2-70b","modelName":"Llama 2-70B","provider":"Meta","rawName":"Llama 2-70B","variant":"epoch","setting":"Epoch AI run, accuracy","score":26.33,"unit":"%","rank":271,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-3-8b","modelName":"Llama 3-8B","provider":"Meta","rawName":"Llama 3-8B","variant":"epoch","setting":"Epoch AI run, accuracy","score":26.07,"unit":"%","rank":272,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"ministral-3b","modelName":"Ministral 3B","provider":"Mistral AI","rawName":"Ministral 3B","variant":"epoch","setting":"Epoch AI run, accuracy","score":25.25,"unit":"%","rank":273,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-llm-67b","modelName":"DeepSeek LLM 67B","provider":"DeepSeek","rawName":"DeepSeek LLM 67B","variant":"epoch","setting":"Epoch AI run, accuracy","score":24.62,"unit":"%","rank":274,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"granite-4-0-1b","modelName":"granite-4.0-1b","provider":"IBM","rawName":"granite-4.0-1b","variant":"epoch","setting":"Epoch AI run, accuracy","score":23.99,"unit":"%","rank":275,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"llama-3-2-1b","modelName":"Llama 3.2 1B","provider":"Meta","rawName":"Llama 3.2 1B","variant":"epoch","setting":"Epoch AI run, accuracy","score":23.93,"unit":"%","rank":276,"rankComputedBySite":true,"date":"2026-08-30","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemma-3-4b","modelName":"Gemma 3 4B","provider":"Google","rawName":"Gemma 3 4B","variant":"epoch","setting":"Epoch AI run, accuracy","score":23.23,"unit":"%","rank":277,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"gemma-3-1b","modelName":"Gemma 3 1B","provider":"Google","rawName":"Gemma 3 1B","variant":"epoch","setting":"Epoch AI run, accuracy","score":19.95,"unit":"%","rank":278,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"mistral-7b-v0-3","modelName":"Mistral 7B v0.3","provider":"Mistral AI","rawName":"Mistral 7B v0.3","variant":"epoch","setting":"Epoch AI run, accuracy","score":15.18,"unit":"%","rank":279,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"yi-34b","modelName":"Yi-34B","provider":"01.AI","rawName":"Yi-34B","variant":"epoch","setting":"Epoch AI run, accuracy","score":14.74,"unit":"%","rank":280,"rankComputedBySite":true,"date":"2025-01-27","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"granite-4-0-350m","modelName":"granite-4.0-350m","provider":"IBM","rawName":"granite-4.0-350m","variant":"epoch","setting":"Epoch AI run, accuracy","score":11.24,"unit":"%","rank":281,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"},{"model":"deepseek-r1-0528-qwen3-8b","modelName":"deepseek-r1-0528-qwen3-8b","provider":"DeepSeek","rawName":"deepseek-r1-0528-qwen3-8b","variant":"epoch","setting":"Epoch AI run, accuracy","score":9.28,"unit":"%","rank":282,"rankComputedBySite":true,"date":"2026-08-28","sourceType":"independent","source":"epoch","sourceUrl":"https://epoch.ai/benchmarks","collected":"2026-10-10"}]}